Hybrid CPU-GPU acceleration of a multithreaded image stitching algorithm

Name: Hybrid CPU-GPU acceleration of a multithreaded image stitching algorithm
Author: Tesfay, Shewıt Weldu, Demirdağ, Zeynep Gülbeyaz, Uğurdağ, Hasan Fatih, Ates, H. F.

İsim	Hybrid CPU-GPU acceleration of a multithreaded image stitching algorithm
Yazar	Tesfay, Shewıt Weldu, Demirdağ, Zeynep Gülbeyaz, Uğurdağ, Hasan Fatih, Ates, H. F.
Basım Tarihi:	2022
Basım Yeri	- IEEE
Konu	CPU-GPU, Image stitching, Multithreading, Real-time image processing
Tür	Belge
Dil	İngilizce
Dijital	Evet
Yazma	Hayır
Kütüphane:	Özyeğin Üniversitesi
Demirbaş Numarası	978-166547010-0
Kayıt Numarası	a7576f77-d8c5-46b7-af3c-b5cb751deba2
Lokasyon	Electrical & Electronics Engineering
Tarih	2022
Notlar	TÜBİTAK
Örnek Metin	Real-time image stitching is critical, especially in un-manned aerial vehicles, and its acceleration has received attention in recent years. This paper describes an image stitching acceleration scheme for heterogeneous (CPU+GPU) devices. Acceleration is attempted with both multithreading and multiprocessing. Most time-critical functions in the algorithm are offloaded on to the GPU. We crafted a 3-buffer ping-pong mechanism for synchro-nization and data transfer among threads/processes in order to maximize CPU utilization. We carried out our experiments on Nvidia Jetson AGX Xavier. Results show that more than 3x acceleration is achieved.
DOI	10.1109/UBMK55850.2022.9919473

Kaynağa git Özyeğin Üniversitesi

Aramaya Dön

Özyeğin Üniversitesi

Kaynağa git

Hybrid CPU-GPU acceleration of a multithreaded image stitching algorithm

Yazar Tesfay, Shewıt Weldu, Demirdağ, Zeynep Gülbeyaz, Uğurdağ, Hasan Fatih, Ates, H. F.

Basım Tarihi 2022

Basım Yeri - IEEE

Konu CPU-GPU, Image stitching, Multithreading, Real-time image processing

Tür Belge

Dil İngilizce

Dijital Evet

Yazma Hayır

Kütüphane Özyeğin Üniversitesi

Demirbaş Numarası 978-166547010-0

Kayıt Numarası a7576f77-d8c5-46b7-af3c-b5cb751deba2

Lokasyon Electrical & Electronics Engineering

Tarih 2022

Notlar TÜBİTAK

Örnek Metin Real-time image stitching is critical, especially in un-manned aerial vehicles, and its acceleration has received attention in recent years. This paper describes an image stitching acceleration scheme for heterogeneous (CPU+GPU) devices. Acceleration is attempted with both multithreading and multiprocessing. Most time-critical functions in the algorithm are offloaded on to the GPU. We crafted a 3-buffer ping-pong mechanism for synchro-nization and data transfer among threads/processes in order to maximize CPU utilization. We carried out our experiments on Nvidia Jetson AGX Xavier. Results show that more than 3x acceleration is achieved.

DOI 10.1109/UBMK55850.2022.9919473